The goal of this proposal is to determine the amount of variation that exists in structural genes coding for specific enzymes and proteins in natural populations in order to test a major prediction of the neutral theory of evolution. This analysis will rely upon a powerful new strategy for determining the DNA sequences of many alleles of the same gene. DNA variation will be quantified in two species, Drosophila melanogaster and D. simulans, for a two-gene region encompassing alcohol dehydrogenase (Adh) and for the white locus (white). We have recently developed a conservative statistical test of the neutral theory of molecular evolution based on a predication of the neutral theory that levels of within-species polymorphism in different regions of DNA should be positively corelated with the corresponding levels of between-species divergence. The data to be collected will allow us to ask for each locus whether patterns of polymorphism and divergence depart from neutral expection and will allow us to test some specific hypotheses to explain an apparent departure from neutrality of the Adh locus and its s' flanking region. The kinds of genetic variation in proteins in natural populations, and the forces of selection operating on them is of direct interest to problems of genetic disease and to variation in response to conditions of the environment.